Skip to content
Agentmart
Submit a tool

writing

Use when drafting, editing, or reviewing prose, Markdown documentation, or instruction text.

Skill by ririnto

IndexedMIT
Write a review

Chalk Labs

Need help deploying writing?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/ririnto-writing

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when drafting, editing, or reviewing prose, Markdown documentation, or instruction text.

Use cases: documents, writing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/writing

What's inside

Skill name
writing
Description
Use when drafting, editing, or reviewing prose, Markdown documentation, or instruction text.
License
MIT
Instructions
83 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Writing

Treat prose and presentation guidance as defaults when the user or target format does not specify another style.

## Reader first

Lead with the answer, completed result, or next useful action.
Put commands, paths, and code beside the action they support.
Number steps when readers must perform actions in a required order.

For direct human work, report state and remaining work when they affect a decision, action, or wait.
For agent-origin messages, follow the active role's communication policy.
Report the observed failure and its effect.
Give a cause and fix when evidence supports them.
When the cause is uncertain, state that limit and the next useful diagnostic step.
Offer a duration only when it helps the reader decide and known assumptions support it.

Keep concise lists complete when omitted details would change an option, conclusion, or action.
Use presentation limits to shape visible output, not analysis, source review, or evidence retention.
Offer a next action only when an unresolved user dependency requires the reader to act.
Remove tangents that do not help complete the requested task.

## Direct prose

Use direct verbs, familiar words, and one term per concept.
Name the actor and use active voice when the actor matters.
Use the factual actor, including a compiler or service, without inventing a human subject.
Remove filler, empty emphasis, vague qualifiers, business jargon, and repetitive conclusions.
Avoid rhetorical contrasts, staged reveals, fragments for emphasis, and invented terms.
Vary sentence length and choose punctuation that suits the requested tone.
Prefer short instruction sentences, splitting dense instructions when their conditions remain clear.

Keep each paragraph focused on one idea.
Keep conditions and exceptions with the actions they qualify.
Preserve technical meaning, authority boundaries, required syntax, and exact quotations when removing repetition.
Keep uncertainty when removing a hedge would invent confidence.
Remove adverbs only when their removal preserves technical behavior and the claim's scope.

## Markdown sentences

Write complete sentences in prose and list items.
Put each sentence on its own source line.
Do not force separate sentences together with semicolons or other punctuation.
Do not replace sentence boundaries to fit several instructions onto one line.
Do not reflow prose to a fixed column width.

Preserve headings, code, metadata, links, and exact quotations in their required syntax.
Use lists for parallel, sequential, or comparative items, without nesting unless the hierarchy is necessary.

## Human-facing documents

For human-facing documents, lead with the action or outcome and use clear headings, compact lists, and short paragraphs.
Check the rendered document for readability after editing it.

## Agent-facing documents

Write agent-facing prompts and instruction-management skills as explicit task contracts.
Before drafting, editing, or reviewing agent-facing prompts or instructions, read [Instruction design](references/instruction-design.md) unless its complete content is already loaded.
Agent-facing documents and prompts are exempt from the human-facing presentation rules and rendered-readability check.
Use LaTeX or Mermaid when notation or a diagram makes the instructions clearer.

## Tables

Use tables for comparisons, mappings, or inputs and outputs.
Keep labels and headers concise, without requiring sentence punctuation.
Write explanatory cells as complete sentences with terminal punctuation.
Keep each explanatory cell to one sentence.
Put additional explanations in separate rows instead of joining sentences within a cell.

For multiple items under one label, leave later first cells blank and prefix descriptions with `-` and a space.
Use that continuation form only when the items need separate rows.
Preserve valid Markdown row syntax instead of breaking a cell across source lines.

## YAML scalars

Use `|-` or `>-` for multiline YAML strings to omit the final line feed.
Use `|-` for literal line breaks and `>-` for source wrapping of one logical line.
Apply these rules to Markdown frontmatter, including memory records.
Preserve literal or folded semantics and validate parsed values, including template strings.

## Final pass

Read the changed prose for filler, repeated rules, sentence boundaries, and detached conditions.
Check table explanations for complete sentences and their required terminal punctuation.
Run the target's applicable Markdown check without changing unrelated files.

From ririnto/workgraph, MIT.

More from ririnto/workgraph

Signals

Stars
0
Forks
0
Open issues
5
Last commit (this folder)
Oct 23, 2026 (8 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from ririnto

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get writing running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying writing?

CallGet help
Cite this page

writing skill. Agent Mart. https://agentmart.dev/skills/ririnto-writing (last updated 2026-10-24)